Llama 2 Finetuning notebook not working

I’m trying to run this colab notebook however the finetuned model does not run any differently to the base model. There seems to be an issue with either the training, the saving of the model, or maybe the loading of the finetuned model. I can’t seem to figure it out.

This notebook is the one linked on the Llama 2 model doc page so I was expecting it to work.